一、方案介绍
OpenTelemetry +Prometheus+ Loki + Tempo + Grafana 是一套现代化、云原生的可观测性解决方案组合，涵盖 Trace（链路追踪）、Log（日志）、Metrics（指标） 三大核心维度，为微服务架构中的应用提供统一的可观测性平台。二、组件介绍 
三、系统架构
四、部署示例应用
4.1应用介绍
https://opentelemetry.io/docs/demo/kubernetes-deployment/
官方为大家写了一个 opentelemetry-demo。
这个项目模拟了一个微服务版本的电子商城，主要包含了以下一些项目：
4.2部署应用
4.2.1获取 charts 包
# helm repo add open-telemetry https://open-telemetry.github.io/opentelemetry-helm-charts
# helm pull open-telemetry/opentelemetry-demo --untar
# cd opentelemetry-demo
# ls
Chart.lock Chart.yaml examples grafana-dashboards README.md UPGRADING.md values.yaml
charts ci flagd products templates values.schema.json4.2.2自定义 charts 包，默认 charts 包集成了opentelemetry-collector、prometheus、grafana、opensearch、jaeger 组件，我们先将其禁用
# vim values.yaml
default:
# List of environment variables applied to all components
env:
- name: OTEL_COLLECTOR_NAME
value: center-collector.opentelemetry.svc
opentelemetry-collector:
enabled: false
jaeger:
enabled: false
prometheus:
enabled: false
grafana:
enabled: false
opensearch:
enabled: false4.2.3安装示例应用
# helm install demo . -f values.yaml
- All services are available via the Frontend proxy: http://localhost:8080
by running these commands:
kubectl --namespace default port-forward svc/frontend-proxy 8080:8080
The following services are available at these paths after the frontend-proxy service is exposed with port forwarding:
Webstore http://localhost:8080/
Jaeger UI http://localhost:8080/jaeger/ui/
Grafana http://localhost:8080/grafana/
Load Generator UI http://localhost:8080/loadgen/
Feature Flags UI http://localhost:8080/feature/
# kubectl get pod
NAME READY STATUS RESTARTS AGE
accounting-79cdcf89df-h8nnc 1/1 Running 0 2m15s
ad-dc6768b6-lvzcq 1/1 Running 0 2m14s
cart-65c89fcdd7-8tcwp 1/1 Running 0 2m15s
checkout-7c45459f67-xvft2 1/1 Running 0 2m13s
currency-65dd8c8f6-pxxbb 1/1 Running 0 2m15s
email-5659b8d84f-9ljr9 1/1 Running 0 2m15s
flagd-57fdd95655-xrmsk 2/2 Running 0 2m14s
fraud-detection-7db9cbbd4d-znxq6 1/1 Running 0 2m15s
frontend-6bd764b6b9-gmstv 1/1 Running 0 2m15s
frontend-proxy-56977d5ddb-cl87k 1/1 Running 0 2m15s
image-provider-54b56c68b8-gdgnv 1/1 Running 0 2m15s
kafka-976bc899f-79vd7 1/1 Running 0 2m14s
load-generator-79dd9d8d58-hcw8c 1/1 Running 0 2m15s
payment-6d9748df64-46zwt 1/1 Running 0 2m15s
product-catalog-658d99b4d4-xpczv 1/1 Running 0 2m13s
quote-5dfbb544f5-6r8gr 1/1 Running 0 2m14s
recommendation-764b6c5cf8-lnkm6 1/1 Running 0 2m14s
shipping-5f65469746-zdr2g 1/1 Running 0 2m15s
valkey-cart-85ccb5db-kr74s 1/1 Running 0 2m15s
# kubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
ad ClusterIP 10.103.72.85 <none> 8080/TCP 2m19s
cart ClusterIP 10.106.118.178 <none> 8080/TCP 2m19s
checkout ClusterIP 10.109.56.238 <none> 8080/TCP 2m19s
currency ClusterIP 10.96.112.137 <none> 8080/TCP 2m19s
email ClusterIP 10.103.214.222 <none> 8080/TCP 2m19s
flagd ClusterIP 10.101.48.231 <none> 8013/TCP,8016/TCP,4000/TCP 2m19s
frontend ClusterIP 10.103.70.199 <none> 8080/TCP 2m19s
frontend-proxy ClusterIP 10.106.13.80 <none> 8080/TCP 2m19s
image-provider ClusterIP 10.109.69.146 <none> 8081/TCP 2m19s
kafka ClusterIP 10.104.9.210 <none> 9092/TCP,9093/TCP 2m19s
kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 176d
load-generator ClusterIP 10.106.97.167 <none> 8089/TCP 2m19s
payment ClusterIP 10.102.143.196 <none> 8080/TCP 2m19s
product-catalog ClusterIP 10.109.219.138 <none> 8080/TCP 2m19s
quote ClusterIP 10.111.139.80 <none> 8080/TCP 2m19s
recommendation ClusterIP 10.97.118.12 <none> 8080/TCP 2m19s
shipping ClusterIP 10.107.102.160 <none> 8080/TCP 2m19s
valkey-cart ClusterIP 10.104.34.233 <none> 6379/TCP 2m19s4.2.4接下来创建 ingress 资源，暴露frontend-proxy 服务 8080 端口
apiVersion: traefik.io/v1alpha1
kind: IngressRoute
metadata:
name: demo
spec:
entryPoints:
- web
routes:
- match: Host(`demo.cuiliangblog.cn`)
kind: Rule
services:
- name: frontend-proxy
port: 80804.2.5创建完 ingress 资源后添加 hosts 解析并访问验证。
4.3配置 Ingress 输出
以 ingress 为例，从 Traefik v2.6 开始，Traefik 原生支持使用 OpenTelemetry 协议导出追踪数据（traces），这使你可以将 Traefik 的 tracing 数据发送到兼容 OTel 的后端 。Traefik 部署可参考文档：https://www.cuiliangblog.cn/detail/section/140101250，
access 配置参考文档：https://doc.traefik.io/traefik/observability/access-logs/#opentelemetry# vim values.yaml
experimental: # 实验性功能配置
otlpLogs: true # 日志导出otlp格式
additionalArguments: # 自定义启动参数
- "--experimental.otlpLogs=true"
- "--accesslog.otlp=true"
- "--accesslog.otlp.grpc=true"
- "--accesslog.otlp.grpc.endpoint=center-collector.opentelemetry.svc:4317"
- "--accesslog.otlp.grpc.insecure=true"
metrics: # 指标
addInternals: true # 追踪内部流量
otlp:
enabled: true # 导出otlp格式
grpc: # 使用grpc协议
endpoint: "center-collector.opentelemetry.svc:4317" # OpenTelemetry地址
insecure: true # 跳过证书
tracing: # 链路追踪
addInternals: true # 追踪内部流量（如重定向）
otlp:
enabled: true # 导出otlp格式
grpc: # 使用grpc协议
endpoint: "center-collector.opentelemetry.svc:4317" # OpenTelemetry地址
insecure: true # 跳过证书五、部署组件
评论 (0)